Method of reordering a queue of write requests

Abstract

In some implementations, a method includes, at a latency reduction system configured to reduce latency in writing data to one or more storage entities that are each configured to store data on an object-basis, receiving a scheduling request associated with a write operation. The method also includes determining a wait period for the write operation, where the wait period is less than or equal to the difference between a client-defined operation threshold and an operation completion time of the write operation and determining an operation start time for the write operation, based in part on the wait period.

What is claimed is: 
     
       1. A method comprising:
 at a latency reduction system to reduce latency in writing data to one or more storage entities that are each configured to store data on an object-basis, the latency reduction system including a non-transitory memory, one or more processors, and control links to each of the one or more storage entities:
 receiving a scheduling request associated with a plurality of write operations; 
 determining a respective wait period for each of the plurality of write operations, wherein each respective wait period is less than or equal to the difference between a write operation threshold and a respective operation completion time of the corresponding write operation; and 
 determining an operation start time for each of the plurality of write operations, based in part on the respective wait period. 
 
 
     
     
       2. The method of  claim 1 , wherein the scheduling request corresponds to a batch request. 
     
     
       3. The method of  claim 1 , wherein the write operation threshold corresponds to a predefined period of time to complete the plurality of write operations. 
     
     
       4. The method of  claim 1 , further comprising:
 determining the respective operation completion time for each of the plurality of write operations at a respective storage entity of the one or more storage entities. 
 
     
     
       5. The method of  claim 1 , wherein the scheduling request is received from an ingest storage entity of the one or more storage entities. 
     
     
       6. The method of  claim 5 , further comprising:
 generating a scheduling response message corresponding to the scheduling request, wherein the scheduling response message includes the operation start time for each of the plurality of write operations; and 
 transmitting the scheduling response message to the ingest storage entity. 
 
     
     
       7. The method of  claim 6 , wherein the scheduling response message identifies a data-storing storage entity or a parity-storing storage entity corresponding to the plurality of write operations. 
     
     
       8. The method of  claim 1 , further comprising:
 obtaining one or more performance parameter values from one or more respective storage entities of the one or more storage entities; and 
 determining the operation start time for each of the plurality of write operations, based in part on the one or more performance parameter values. 
 
     
     
       9. The method of  claim 1 , further comprising:
 selecting a data-storing storage entity from the one or more storage entities for writing content data corresponding to the plurality of write operations, when at least one of the plurality of write operations include an instruction to write content data; 
 selecting a parity-storing storage entity from the one or more storage entities for writing parity data corresponding to at least one of the plurality of write operations, when at least one of the plurality of write operations include an instruction to write parity data; and 
 determining the operation start time for each of the plurality of write operations, based in part on the data-storing storage entity or the parity-storing storage entity. 
 
     
     
       10. The method of  claim 1 , wherein the scheduling request includes one or more of a data segment, a location for retrieving the data segment, a size of the data segment, a copy count of the data segment, the write operation threshold, or a data-to-parity ratio for storage and backup in the one or more storage entities. 
     
     
       11. A latency reduction system to reduce latency in writing data to one or more storage entities, comprising:
 one or more storage entities each configured to store data on an object-basis; and 
 a controller with control links to the one or more storage entities, comprising:
 a communication module to receive a scheduling request associated with a plurality of write operations; and 
 a scheduling module to: 
 
 determine a respective wait period for each of the plurality of write operations, wherein each respective wait period is less than or equal to the difference between a write operation threshold and a respective operation completion time of the corresponding write operation; and 
 determine an operation start time for each of the plurality of write operations, based in part on the respective wait period. 
 
     
     
       12. The latency reduction system of  claim 11 , wherein the scheduling request corresponds to a batch request. 
     
     
       13. The latency request system of  claim 11 , wherein the write operation threshold corresponds to a predefined period of time to complete the plurality of write operations. 
     
     
       14. The latency reduction system of  claim 11 , wherein the scheduling module is further configured to:
 obtain one or more performance parameter values from one or more respective storage entities of the one or more storage entities; and 
 determine the operation start time for each of the plurality of write operations, based in part on the one or more performance parameter values. 
 
     
     
       15. The latency reduction system of  claim 11 , wherein the scheduling module is further configured to:
 select a data-storing storage entity from the one or more storage entities for writing content data corresponding to the plurality of write operations, when at least one of the plurality of write operations include an instruction to write content data; 
 select a parity-storing storage entity from the one or more storage entities for writing parity data corresponding to at least one of the plurality of write operations, when at least one of the plurality of write operations include an instruction to write parity data; and 
 determine the operation start time for each of the plurality of write operations, based in part on the data-storing storage entity or the parity-storing storage entity. 
 
     
     
       16. A latency reduction device configured to reduce latency in writing data to one or more storage entities each configured to store data on an object-basis, comprising:
 one or more processors; 
 a non-transitory memory; 
 one or more control links to the one or more storage entities; 
 means for receiving a scheduling request associated with a plurality of write operations; 
 means for determining a respective wait period for each of the plurality of write operations, wherein each respective wait period is less than or equal to the difference between a write operation threshold and a respective operation completion time of the corresponding write operation; and 
 means for determining an operation start time for each of the plurality of write operations, based in part on the respective wait period. 
 
     
     
       17. The latency reduction device of  claim 16 , wherein the scheduling request corresponds to a batch request. 
     
     
       18. The latency reduction device of  claim 16 , wherein the write operation threshold corresponds to a predefined period of time to complete the plurality of write operations. 
     
     
       19. The latency reduction device of  claim 16 , further comprising:
 means for obtaining one or more performance parameter values from one or more respective storage entities of the one or more storage entities; and 
 means for determining the operation start time for each of the plurality of write operations, based in part on the one or more performance parameter values. 
 
     
     
       20. The latency reduction device of  claim 16 , further comprising:
 means for selecting a data-storing storage entity from the one or more storage entities for writing content data corresponding to the plurality of write operations, when at least one of the plurality of write operations include an instruction to write content data; 
 means for selecting a parity-storing storage entity from the one or more storage entities for writing parity data corresponding to at least one of the plurality of write operations, when at least one of the plurality of write operations include an instruction to write parity data; and 
 means for determining the operation start time for each of the plurality of write operations, based in part on the data-storing storage entity or the parity-storing storage entity.